Method and system for discovering commuter traffic cells based on inter-regional functional attributes and transfer passenger flow

Abstract

The present application relates to a kind of commuting traffic cell pair discovery method and system based on interregional function attribute and transfer passenger flow, belong to intelligent transportation technical field.The method includes the following steps: S1, according to city road network grade, traffic cell is divided, constructs traffic cell network;S2, in combination with city operation vehicle data and GIS technology, the OD point in data is mapped to inside traffic cell, forms OD transfer matrix;S3, using city POI data, in combination with kernel density estimation and POI fixed weight, the function attribute of each traffic cell is identified;S4, by city operation vehicle OD transfer matrix, the flow fluctuation rate between regions is calculated, and in combination with traffic cell function attribute, commuting traffic cell pair with significant commuting relationship is found.The present application is found by using city operation vehicle data to reveal the commuting traffic cell pair of city, is favorable to reveal the commuting flow transfer relationship between city regions, is helpful to provide guidance and basis for the layout and adjustment of city traffic service.

Description

Technical Field

[0001] This invention belongs to the field of intelligent transportation technology and relates to a method and system for discovering commuter traffic zones based on inter-regional functional attributes and passenger flow transfers. Background Technology

[0002] With the continuous development of urbanization in China, the expansion of urban areas, and the increasing urban population density, significant changes have occurred in the urban spatial structure. Among these changes, the separation of work and residence has gradually impacted the daily lives of urban residents, altering traditional commuting methods. Traditionally, urban residents relied mainly on public transportation, rail, and private cars for commuting. However, with the increasing demand for customized travel, commuting using urban commercial vehicles (ride-hailing and taxis) has become increasingly popular. According to the "2022 Travel Consumption Report" released by T3 Mobility, commuting to and from work has become the primary reason for taking taxis, and among many cities, Chongqing passengers are the most frequent users of taxis for commuting. Chongqing is a mountainous city with rugged and complex terrain, significant elevation differences, and numerous bridges and tunnels. The complex and narrow urban roads, coupled with continuous population growth and increasing car ownership, have resulted in severe traffic congestion in Chongqing.

[0003] Thanks to the booming development of the big data era, we now have sufficient data support for exploring commuting patterns between urban areas. The vast amounts of data collected by various data acquisition devices throughout the city provide the foundation for understanding urban traffic conditions and building "smart transportation." How to use data mining techniques to glean insights into commuting patterns between urban areas from this massive dataset is of paramount importance in solving urban traffic congestion. Urban operating vehicle data, as a crucial component of urban traffic big data, holds significant research value. Discovering urban commuting traffic zones using urban operating vehicle data helps reveal the relationships between commuting flow transfers between urban areas, providing guidance and a basis for the layout and adjustment of urban transportation services. Summary of the Invention

[0004] In view of this, the purpose of the present invention is to provide a method and system for discovering commuter traffic zone pairs based on inter-regional functional attributes and transferred passenger flow. The method mainly achieves intelligent management of urban commuting by dividing traffic zones, obtaining OD transfer matrices, identifying zone functional attributes, and discovering commuter traffic zone pairs. This method is conducive to revealing the commuter flow transfer relationship between urban areas and helps to provide guidance and basis for the layout and adjustment of urban transportation services.

[0005] To achieve the above objectives, the present invention provides the following technical solution:

[0006] A method for discovering commuter traffic zones based on inter-regional functional attributes and passenger flow includes the following steps:

[0007] S1. Divide traffic zones according to the urban road network level and construct a traffic zone network;

[0008] S2. By combining urban operating vehicle data with GIS (Geographic Information System) technology, the OD points in the data are mapped to the inside of traffic zones to form an OD transfer matrix;

[0009] S3. Using urban POI data, combined with kernel density estimation and POI weighting, identify the functional attributes of each traffic zone;

[0010] S4. By using the OD transfer matrix of urban operating vehicles, the inter-regional traffic fluctuation rate is calculated, and combined with the functional attributes of traffic zones, commuter traffic zone pairs with significant commuting relationships are identified.

[0011] Furthermore, step S1 specifically includes:

[0012] Step S11: Obtain SHP road network data: Download the ESRI Shapefile, an open spatial data format file, through the OSM road network open source data platform;

[0013] Step S12: Trim the study area: In ArcGIS software, import the road network file and trim it to determine the study area;

[0014] Step S13: Select road level: Different levels of road network divide traffic zones of varying sizes and levels of detail.

[0015] Step S14: Road network data to raster: Use the feature to raster tool in ArcGIS to convert the main road network data into raster. During this process, it is necessary to set the output cell size, which will directly affect the division result.

[0016] Step S15: Grid settings: After obtaining the road network grid, the grid needs to be set, with the road set to 1 and the background set to 0;

[0017] Step S16: Dilation and Dilution: Export the raster set up in the previous step and perform dilation and dilution operations using PyCharm software. The purpose of dilation is to remove some unnecessary details in the map and prevent these details from forming connected areas. However, since the dilation operation may cause urban road segments to expand, a dilution operation is needed to extract the backbone of the road network while maintaining the topology of the original binary image.

[0018] Step S17: Generate traffic zones: Import the dilated and diluted raster image into ArcGIS, use the raster to polygon tool to convert it into a shapefile, and obtain the traffic zones after division.

[0019] Furthermore, step S2 specifically includes:

[0020] Step S21: Pruning: Because the order data in the urban operating vehicle dataset is not entirely within the scope of the study, it is necessary to remove order data where points O or D are outside the scope of the study.

[0021] Step S22: Mapping: Although the point data points O and D retained by cropping are within the research scope, they have not been associated with the cell. In order to establish an association with the cell, it is necessary to use ArcGIS's spatial connection function to map the point data to the cell and associate each point data with its cell.

[0022] Step S23: Connection: After mapping, each point data will have a new attribute added, namely the cell number. However, these data are only OD point data and need to be converted back into order data. To do this, the OD point data need to be connected by the order number to obtain order data containing information about the cell where point O is located and the cell where point D is located.

[0023] Step S24: Formation: Use pivot tables to convert the processed order data into an OD transition matrix.

[0024] Furthermore, step S3 specifically includes:

[0025] Step S31: POI Data Preprocessing: The original POI data is diverse, and some locations are meaningless, such as parking lot entrances and ATMs. Furthermore, some data is duplicated. Therefore, the data first needs to be deduplicated. After deduplication, the POI data is reclassified according to the "Urban Land Use Classification and Planning Construction Land Standard GBJ 137-90" and the "Research on POI Classification Standards" document.

[0026] Step S32: Kernel Density Estimation: Calculate the kernel density estimate for each type of POI within the traffic zone. The calculation formula is as follows:

[0027]

[0028] In the formula, f(x) refers to the kernel density estimate at position x; x i refers to the i-th POI; h is the attenuation value, also known as bandwidth; n refers to the number of POIs whose distance from x is no greater than h; k refers to the pre-defined kernel function;

[0029] The kernel density estimation results are affected by the kernel function and the attenuation value. The kernel function has a relatively small impact on the results, while the attenuation value has a more significant impact. Therefore, the actual situation needs to be considered when selecting the attenuation value, and an appropriate attenuation value should be chosen. After multiple calculations and experimental analyses, this invention sets the attenuation value to 1500 meters. The formula for calculating the attenuation value h is as follows:

[0030]

[0031] In the formula, D m SD refers to the median distance between the mean center point and all points; N refers to the total number of sample points; SD refers to the standard distance between the mean center point and all points, and its calculation formula is as follows:

[0032]

[0033] In the formula, x i and y i This refers to the coordinates of the i-th POI. and This refers to the average center point coordinates of this type of POI.

[0034] Step S33: POI Weighting: Since each POI is merely a coordinate point abstracted from a geographic entity, and different categories of POIs differ not only in area but also in public awareness and urban landmark influence, it is necessary to classify and weight each type of POI. This invention references existing data to set influence weights and area weights for each type of POI, and comprehensively obtains a unique weight value for each type of POI.

[0035] Step S34: Quantitative Identification: For each traffic zone, calculate the weighted values ​​of various functional attribute POIs within it and summarize them; by calculating the proportion of the weighted values ​​of various POIs within the traffic zone, the functional attributes of each traffic zone are identified, and its spatial characteristics are explored. The calculation formula is as follows:

[0036]

[0037] In the formula, C ij N represents the weight score of the j-th type of POI within the i-th traffic zone, meaning the proportion of the weighted value of the j-th type of POI within the i-th traffic zone to the total weighted values ​​of all types of POIs within that traffic zone; j W represents the kernel density value of the j-th type of POI within the traffic zone; j The weight of POI of type j is represented by m; the number of all POI types in the traffic cell is represented by m; k represents POI of type k; N k W represents the kernel density value of the k-th type of POI within the traffic zone;k This represents the weight of the k-th type of POI. Using the above formula, the weight scores of different POI categories in each traffic zone can be calculated. If the weight score of a certain type of POI exceeds 50%, the traffic zone is considered a single-function zone, and its functional attribute is that type of POI. If no type of POI has a weight score exceeding 50%, the traffic zone is determined to be a mixed-function zone, and its functional attribute is a mixture of the top two POIs with the highest weight scores.

[0038] Furthermore, step S4 specifically includes:

[0039] Step S41: Discover traffic flow stability pairs: By calculating the coefficient of variation (COD) and the inter-regional traffic flow fluctuation range (RFR) between each traffic pair, the inter-regional traffic flow fluctuation rate (RFF) is obtained. If the traffic flow fluctuation rate (RFF) between the pairs is less than the set threshold maxRFF, then the traffic pair is determined to be a traffic flow stability pair.

[0040] The formula for calculating the coefficient of variation (COD) is as follows:

[0041]

[0042] In the formula, COD ij Represents the coefficient of variation between traffic cell i and traffic cell j; PassengerFlow ij This represents the passenger flow between traffic zone i and traffic zone j in the OD transition matrix; Mean represents the mean; StandardDeviation represents the standard deviation.

[0043] The formula for calculating the inter-regional flow fluctuation range (RFR) is as follows:

[0044]

[0045] In the formula, RFR ij This represents the range of inter-regional traffic flow fluctuations between traffic cell i and traffic cell j; PassengerFlow ij This represents the passenger flow between traffic zone i and traffic zone j in the OD transition matrix; max represents the maximum value; min represents the minimum value.

[0046] The formula for calculating the inter-regional flow volatility (RFF) is as follows:

[0047]

[0048] In the formula, RFF ij This represents the inter-regional traffic flow fluctuation rate between traffic zone i and traffic zone j; COD ijRepresents the coefficient of variation between traffic cell i and traffic cell j; RFR ij This represents the range of traffic flow fluctuations between traffic zone i and traffic zone j.

[0049] Step S42: Discover commuter neighborhood pairs: A commuter neighborhood pair is a neighborhood pair with stable passenger flow that has commuting characteristics. That is, the geospatial characteristics of the commuter neighborhood pair must meet the commuting OD characteristics, and the flow between the commuter neighborhood pairs must remain stable during the morning and evening peak hours. Specifically, if a traffic neighborhood pair consists of neighborhood pairs with stable passenger flow during the morning and evening peak hours, and their respective spatial functional attributes meet the commuting characteristics, then this neighborhood pair is determined to be a commuter neighborhood pair.

[0050] The beneficial effects of this invention are as follows:

[0051] The technical solution of this invention discovers commuter traffic zone pairs by dividing traffic zones, obtaining OD transfer matrices, and identifying zone functional attributes. By utilizing urban operating vehicle data to discover urban commuter traffic zone pairs, it is beneficial to reveal the commuter traffic transfer relationship between urban areas and to provide guidance and basis for the layout and adjustment of urban transportation services. This invention has broad application prospects.

[0052] Other advantages, objectives, and features of the invention will be set forth in part in the description which follows, and in part will be apparent to those skilled in the art from the following examination, or may be learned from practice of the invention. The objectives and other advantages of the invention can be realized and obtained through the following description. Attached Figure Description

[0053] To make the objectives, technical solutions, and advantages of the present invention clearer, the preferred embodiments of the present invention will be described in detail below with reference to the accompanying drawings, wherein:

[0054] Figure 1 This is a general framework diagram of the present invention;

[0055] Figure 2 This is a schematic diagram of traffic zone division according to the present invention;

[0056] Figure 3 This is a schematic diagram illustrating the functional attribute identification of traffic zones according to the present invention;

[0057] Figure 4 This is a schematic diagram illustrating the commuter traffic community detection method of the present invention. Detailed Implementation

[0058] The technical solution of the present invention will now be described in detail with reference to the accompanying drawings.

[0059] Figure 1As shown in the figure, the present invention provides a method for discovering commuter traffic zones based on inter-regional functional attributes and shifting passenger flow, which mainly includes the following steps:

[0060] S1. Divide traffic zones according to the urban road network level and construct a traffic zone network; Figure 2 This is a schematic diagram of traffic zone division according to the present invention.

[0061] S2. By combining urban operating vehicle data with GIS (Geographic Information System) technology, the OD points in the data are mapped to the inside of traffic zones to form an OD transfer matrix;

[0062] S3. Using urban POI data, combined with kernel density estimation and POI weighting, identify the functional attributes of each traffic zone; Figure 3 This is a schematic diagram illustrating the functional attribute identification of traffic zones according to the present invention.

[0063] S4. By using the OD transfer matrix of urban operating vehicles, the inter-regional traffic fluctuation rate is calculated, and combined with the functional attributes of traffic zones, commuter traffic zone pairs with significant commuting relationships are identified.

[0064] In this embodiment, step S1 specifically includes:

[0065] Step S11: Obtain SHP road network data: Download all spatial data open format files (ESRIShapefile) for Chongqing Municipality through the OSM road network open source data platform;

[0066] Table 1 shows the data format for urban operating vehicles in Chongqing:

[0067] Table 1 Data Format Table for Urban Operating Vehicles in Chongqing

[0068]

[0069] Step S12: Crop the research scope: In ArcGIS software, import the road network file and crop it to determine the research scope; since the ride-hailing dataset and taxi dataset used in this invention are mostly located in the main urban area of ​​Chongqing, the research scope is limited to the nine main urban areas of Chongqing, including Yuzhong District, Dadukou District, Jiangbei District, Shapingba District, Jiulongpo District, Nan'an District, Beibei District, Yubei District and Banan District.

[0070] Step S13: Select road level: The traffic zones divided by different levels of road networks are of different sizes and have different levels of precision; select the main roads (5112, 5113) in the urban road network as the roads for dividing the traffic zones in this invention.

[0071] Table 2 is a road classification table:

[0072] Table 2 Road Classification Table

[0073]

[0074]

[0075] Step S14: Road Network Data to Raster: Using the feature-to-raster tool in ArcGIS, the road network data of the main roads is converted into a raster. During this process, the output cell size needs to be set, as the cell size directly affects the subdivision results; for example, if the cell size is set too large, adjacent roads will be merged together. In this invention, the cell size is set to 2e-04, and the number of generated subdivisions is 237.

[0076] Step S15: Grid settings: After obtaining the road network grid, the grid needs to be set, with the road set to 1 and the background set to 0;

[0077] Step S16: Dilation and Dilution: Export the raster set up in the previous step and perform dilation and dilution operations using PyCharm software. The purpose of dilation is to remove some unnecessary details in the map and prevent these details from forming connected areas. However, since the dilation operation may cause urban road segments to expand, a dilution operation is needed to extract the backbone of the road network while maintaining the topology of the original binary image.

[0078] Step S17: Generate traffic zones: Import the dilated and diluted raster image into ArcGIS, use the raster to polygon tool to convert it into a shapefile, and obtain the traffic zones after division.

[0079] Step S2 specifically includes:

[0080] Step S21: Pruning: Because the order data in the urban operating vehicle dataset is not entirely within the scope of the study, it is necessary to remove order data where points O or D are outside the scope of the study.

[0081] Step S22: Mapping: Although the point data points O and D retained by cropping are within the research scope, they have not been associated with the cell. In order to establish an association with the cell, it is necessary to use ArcGIS's spatial connection function to map the point data to the cell and associate each point data with its cell.

[0082] Step S23: Connection: After mapping, each point data will have a new attribute added, namely the cell number. However, these data are only OD point data and need to be converted back into order data. To do this, the OD point data need to be connected by the order number to obtain order data containing information about the cell where point O is located and the cell where point D is located.

[0083] Step S24: Formation: Use pivot tables to convert the processed order data into an OD transition matrix.

[0084] Step S3 specifically includes:

[0085] Step S31: POI Data Preprocessing: This invention selects approximately 1.16 million POI data entries from Gaode Maps within Chongqing. The data format mainly includes address, name, administrative region, postal code, longitude, latitude, and data type. The original POI data types include 28 major categories (mainly covering scenic spots, catering services, shopping services, living services, commercial and residential, accommodation services, etc.), 209 medium categories, and 1388 minor categories. The original POI data is diverse and contains some locations with no practical significance, such as parking lot entrances and ATMs. Furthermore, some data is duplicated. Therefore, the data first needs to be deduplicated. After deduplication, the deduplicated POI data is reclassified with reference to the "Urban Land Use Classification and Planning Construction Land Standard GBJ 137-90" and the "Research on POI Classification Standards" document.

[0086] Table 3 is a classification table of POI data:

[0087] Table 3 POI Data Classification Table

[0088]

[0089] Step S32: Kernel Density Estimation: Calculate the kernel density estimate for each type of POI within the traffic zone. The calculation formula is as follows:

[0090]

[0091] In the formula, f(x) refers to the kernel density estimate at position x; x i refers to the i-th POI; h is the attenuation value, also known as bandwidth; n refers to the number of POIs whose distance from x is no greater than h; k refers to the pre-defined kernel function;

[0092] The kernel density estimation results are affected by the kernel function and the attenuation value. The kernel function has a relatively small impact on the results, while the attenuation value has a more significant impact. Therefore, the actual situation needs to be considered when selecting the attenuation value, and an appropriate attenuation value should be chosen. After multiple calculations and experimental analyses, this invention sets the attenuation value to 1500 meters. The formula for calculating the attenuation value h is as follows:

[0093]

[0094] In the formula, D m SD refers to the median distance between the mean center point and all points; N refers to the total number of sample points; SD refers to the standard distance between the mean center point and all points, and its calculation formula is as follows:

[0095]

[0096] In the formula, x i and y i This refers to the coordinates of the i-th POI. and This refers to the average center point coordinates of this type of POI.

[0097] Step S33: POI Weighting: Since each POI is merely a coordinate point abstracted from a geographic entity, and different categories of POIs differ not only in area but also in public awareness and urban landmark influence, it is necessary to classify and weight each type of POI. This invention references existing data to set influence weights and area weights for each type of POI, and comprehensively obtains a unique weight value for each type of POI.

[0098] Table 4 shows the POI weight table:

[0099] Table 4 POI Weight Table

[0100]

[0101]

[0102] Step S34: Quantitative Identification: For each traffic zone, calculate the weighted values ​​of various functional attribute POIs within it and summarize them; by calculating the proportion of the weighted values ​​of various POIs within the traffic zone, the functional attributes of each traffic zone are identified, and its spatial characteristics are explored. The calculation formula is as follows:

[0103]

[0104] In the formula, C ijN represents the weight score of the j-th type of POI within the i-th traffic zone, meaning the proportion of the weighted value of the j-th type of POI within the i-th traffic zone to the total weighted values ​​of all types of POIs within that traffic zone; j W represents the kernel density value of the j-th type of POI within the traffic zone; j The weight of POI of type j is represented by m; the number of all POI types in the traffic cell is represented by m; k represents POI of type k; N k W represents the kernel density value of the k-th type of POI within the traffic zone; k This represents the weight of the k-th type of POI. Using the above formula, the weight scores of different POI categories in each traffic zone can be calculated. If the weight score of a certain type of POI exceeds 50%, the traffic zone is considered a single-function zone, and its functional attribute is that type of POI. If no type of POI has a weight score exceeding 50%, the traffic zone is determined to be a mixed-function zone, and its functional attribute is a mixture of the top two POIs with the highest weight scores.

[0105] Step S4 specifically includes:

[0106] Step S41: Discover traffic flow stability pairs: By calculating the coefficient of variation (COD) and the inter-regional traffic flow fluctuation range (RFR) between each traffic pair, the inter-regional traffic flow fluctuation rate (RFF) is obtained. If the traffic flow fluctuation rate (RFF) between the pairs is less than the set threshold maxRFF, then the traffic pair is determined to be a traffic flow stability pair.

[0107] The formula for calculating the coefficient of variation (COD) is as follows:

[0108]

[0109] In the formula, COD ij Represents the coefficient of variation between traffic cell i and traffic cell j; PassengerFlow ij This represents the passenger flow between traffic zone i and traffic zone j in the OD transition matrix; Mean represents the mean; and standard deviation represents the standard deviation.

[0110] The formula for calculating the inter-regional flow fluctuation range (RFR) is as follows:

[0111]

[0112] In the formula, RFR ij This represents the range of inter-regional traffic flow fluctuations between traffic cell i and traffic cell j; PassengerFlow ij This represents the passenger flow between traffic zone i and traffic zone j in the OD transition matrix; max represents the maximum value; min represents the minimum value.

[0113] The formula for calculating the inter-regional flow volatility (RFF) is as follows:

[0114]

[0115] In the formula, RFF ij This represents the inter-regional traffic flow fluctuation rate between traffic zone i and traffic zone j; COD ij Represents the coefficient of variation between traffic cell i and traffic cell j; RFR ij This represents the range of traffic flow fluctuations between traffic zones i and j. In the ride-hailing and taxi datasets, the proportion of RFF values ​​between 0 and 0.3 during both morning and evening peak hours exceeds 70%, indicating that the traffic flow fluctuation level between zones is relatively stable within the RFF < 0.3 range. Therefore, drawing on relevant statistical criteria, this invention selects maxRFF = 0.3.

[0116] Step S42: Discover commuter neighborhood pairs: A commuter neighborhood pair is a neighborhood pair with stable passenger flow that has commuting characteristics. That is, the geospatial characteristics of the commuter neighborhood pair must meet the commuting OD characteristics, and the flow between the commuter neighborhood pairs must remain stable during the morning and evening peak hours. Specifically, if a traffic neighborhood pair consists of neighborhood pairs with stable passenger flow during the morning and evening peak hours, and their respective spatial functional attributes meet the commuting characteristics, then this neighborhood pair is determined to be a commuter neighborhood pair. Figure 4 This is a schematic diagram illustrating the commuter traffic community detection method of the present invention.

[0117] In summary, the technical solution of this invention mainly achieves intelligent management of urban commuting by dividing traffic zones, obtaining the OD transfer matrix, identifying the functional attributes of the zones, and discovering commuter traffic zone pairs. This helps to reveal the commuter flow transfer relationship between urban areas and provides guidance and basis for the layout and adjustment of urban transportation services.

[0118] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications can be made to the technical solutions of the present invention without departing from the spirit and scope of the present invention, and all such modifications should be covered within the scope of the claims of the present invention.

Claims

1. A method for discovering commuter traffic zones based on inter-regional functional attributes and shifting passenger flow, characterized in that: The method includes the following steps: S1. Divide traffic zones according to the urban road network level and construct a traffic zone network; S2. By combining urban operating vehicle data with GIS technology, the OD points in the data are mapped to the inside of traffic zones to form an OD transfer matrix; S3. Using urban POI data, combined with kernel density estimation and POI weighting, identify each traffic point. Functional attributes of the area; S4. Using the OD transfer matrix of urban operating vehicles, the inter-regional traffic flow fluctuation rate is calculated. Combined with the functional attributes of traffic zones, commuter traffic zone pairs with significant commuting relationships are identified, specifically including: Step S41: Discover traffic flow stability pairs: By calculating the coefficient of variation (COD) and the inter-regional traffic flow fluctuation range (RFR) between each traffic pair, the inter-regional traffic flow fluctuation rate (RFF) is obtained. If the traffic flow fluctuation rate (RFF) between the pairs is less than the set threshold maxRFF, then the traffic pair is determined to be a traffic flow stability pair. The formula for calculating the coefficient of variation (COD) is as follows: In the formula, Indicates traffic community With traffic community The coefficients of variation between them; In the OD transition matrix, traffic zones are represented. With traffic community Passenger flow between; This represents the mean; Indicates standard deviation; The formula for calculating the inter-regional flow fluctuation range (RFR) is as follows: In the formula, Indicates traffic community With traffic community The range of flow fluctuations between regions; In the OD transition matrix, traffic zones are represented. With traffic community Passenger flow between; Indicates the maximum value; Indicates the minimum value; The formula for calculating the inter-regional flow volatility (RFF) is as follows: In the formula, Indicates traffic community With traffic community Inter-regional flow fluctuation rate; Indicates traffic community With traffic community The coefficients of variation between them; Indicates traffic community With traffic community The range of flow fluctuations between regions; Step S42: Discover commuter neighborhood pairs: A commuter neighborhood pair is a neighborhood pair with stable passenger flow that has commuting characteristics. That is, the geospatial characteristics of the commuter neighborhood pair must meet the commuting OD characteristics, and the flow between the commuter neighborhood pairs must remain stable during the morning and evening peak hours. Specifically, if a traffic neighborhood pair consists of neighborhood pairs with stable passenger flow during the morning and evening peak hours, and their respective spatial functional attributes meet the commuting characteristics, then this neighborhood pair is determined to be a commuter neighborhood pair.

2. The method for discovering commuter traffic zones based on inter-regional functional attributes and passenger flow according to claim 1, characterized in that: Step S1 specifically includes: Step S11: Obtain SHP road network data: Download the ESRI Shapefile, an open spatial data format file, through the OSM road network open source data platform; Step S12: Trim the study area: In ArcGIS software, import the road network file and trim it to determine the study area; Step S13: Select road level: Different levels of road network divide the traffic zones into different sizes and with different levels of detail; Step S14: Road network data to raster: Use the feature to raster tool in ArcGIS to convert the main road network data into raster. During this process, it is necessary to set the output cell size, which will directly affect the division result. Step S15: Grid settings: After obtaining the road network grid, the grid needs to be set, with the road set to 1 and the background set to 0; Step S16: Diffusion and Dilution: Export the raster set up in the previous step and perform diffusion and dilution operations using PyCharm software; Step S17: Generate traffic zones: Import the dilated and diluted raster image into ArcGIS, use the raster to polygon tool to convert it into a shapefile, and obtain the traffic zones after division.

3. The method for discovering commuter traffic zones based on inter-regional functional attributes and passenger flow according to claim 2, characterized in that: Step S2 specifically includes: Step S21: Crop: Remove order data where point O or point D is outside the scope of the study; Step S22: Mapping: Although the point data points O and D retained by cropping are within the research scope, they have not been associated with the cell. In order to establish an association with the cell, it is necessary to use ArcGIS's spatial connection function to map the point data to the cell and associate each point data with its cell. Step S23: Connection: After mapping, each point data will have a new attribute added, namely the cell number. However, these data are only OD point data and need to be converted back into order data. The OD point data are connected by the order number to obtain order data containing information about the cell where point O is located and the cell where point D is located. Step S24: Formation: Use pivot tables to convert the processed order data into an OD transition matrix.

4. The method for discovering commuter traffic zones based on inter-regional functional attributes and passenger flow according to claim 3, characterized in that: Step S3 specifically includes: Step S31: POI data preprocessing: The original POI data is diverse and contains some points that are meaningless. First, the data needs to be deduplicated. After deduplication, the deduplicated POI data is reclassified with reference to the "Urban Land Use Classification and Planning Construction Land Standard GBJ 137-90" and the "Research on POI Classification Standards" document. Step S32: Kernel Density Estimation: Calculate the kernel density estimate for each type of POI within the traffic zone. The calculation formula is as follows: In the formula, This refers to the kernel density estimate at position x; It refers to the first One POI point; This is the attenuation value, also known as bandwidth; It refers to and The distance is no higher than POI points; This refers to a pre-defined kernel function; The results of kernel density estimation are affected by the kernel function and the decay value. The formula for calculating the decay value h is shown below: In the formula, It refers to the median distance between the average center point and all points; This refers to the total number of sample points; This refers to the standard distance between the average center of a given point and all other points, and its calculation formula is as follows: In the formula, and It refers to the first The coordinates of the POI points; and This refers to the average center point coordinates of this type of POI. Step S33: POI weighting: It is necessary to classify and weight each type of POI, set influence weight and area weight for each type of POI, and obtain a unique weight value for each type of POI. Step S34: Quantitative Identification: For each traffic zone, calculate the weighted values ​​of various functional attribute POIs within it and summarize them; by calculating the proportion of the weighted values ​​of various POIs within the traffic zone, the functional attributes of each traffic zone are identified, and its spatial characteristics are explored. The calculation formula is as follows: In the formula, Indicates the first Within the traffic zone, the first The weight score of a class of POI means that the first... Within the traffic zone, the first The weighted value of a POI category relative to the total weighted values ​​of all POI categories within the traffic zone; This indicates the first [unit / item] within the traffic community. Kernel density values ​​for POI-like objects; Indicates the first Weights of POIs; This indicates the total number of all POI categories present within the traffic zone; Indicates the first POI-like; This indicates the first [unit / item] within the traffic community. Kernel density values ​​for POI-like objects; Indicates the first The weight of POI categories is calculated using the above formula. If the weight score of a certain type of POI exceeds 50%, the traffic area is considered a single-function area, and its functional attribute is that type of POI. If no type of POI has a weight score exceeding 50%, the traffic area is determined to be a mixed-function area, and its functional attribute is a mixture of the top two POIs with the highest weight scores.

5. A commuter traffic neighborhood detection system based on inter-regional functional attributes and shifting passenger flow, characterized in that: The system employs the method described in any one of claims 1 to 4.

Images